February 11, 2002

 

Ms. Pam Jarnecke

Bureau of Land Management

Battle Mountain Field Office

50 Bastian Road

Battle Mountain, NV 89820

 

Sent VIA Certified Mail Number

 

Re: Phoenix Project Final EIS

 

Dear Ms. Jarnecke:

 

Thank you for this opportunity to review the final EIS for this document.  We are very disappointed that the BLM did not take the advice of the EPA and redo the draft EIS.  The DEIS had been an extremely poorly written document that raised more questions than it answered.  In over seven years of reviewing NEPA documents, I have never reviewed a worse analysis.  If our concerns are not considered, we strongly second the EPA’s motion in their letter to refer this project to the Council on Environmental Quality.  This comment letter is postmarked by February 11, 2002 and is therefore timely.

 

Especially since the BLM and Newmont essentially ignored them, all comments we provided on the DEIS are reiterated and incorporated here by reference.  Some of those comments and the BLM’s responses are discussed herein.  This letter also includes additional concerns that have come to light since the DEIS.

 

As proposed, this project will cause undue or unnecessary degradation to public lands.  Among many things, this project will pollute groundwater and surface water and likely render the site to be listed for Superfund National Priority Listing.  It will illegally dry streams.  It causes substantial toxicological threats to wildlife.  And it grossly underestimates the needed reclamation bond and trust fund.  These items, and more will be discussed in detail below.

 

Based on our review of the FEIS, the BLM has ignored comments from both the public and their sister agencies.  The EPA requested a revised draft because the document was insufficient.  The BLM refused by just stating that they disagree with the BLM’s analysis.

 

Our main concern with this site is that it be cleaned up and not turn into a Superfund clean-up site as both we and the EPA indicated in comment letters. The BLM has been selling this project as a remining project.  It is not.  Moving a billion tons of rock and deepening pits by hundreds of feet is not necessary to clean the project.  However, we are not opposed to a mine at this site if Newmont is able to mine while cleaning the site with a guarantee that they will not make it worse.  But, the EPA was correct, however, in arguing that Newmont is responsible for cleaning the site whether they mine or not.  Both EPA and GBMW questioned whether the proposal was too large.

 

class=Section2>

EPA believes that BMG/Newmont is responsible for cleaning up contaminated groundwater or surface water resulting from the existing project independent of the proposed project; and that a massive expansion of the project, which would degrade water quality further, is not justified simply because current permits do not require BMG to conduct proper closure and/or post-closure activities at the existing mine.  Comment 1-7, EPA DEIS letter at 5, emphases in original.

 

Our conclusion after reading the FEIS is that this project as proposed unacceptably degrades the groundwater and depletes the surface of the site.  The only alternative that will present acceptable environmental risk is one that provides for complete clean-up of existing contamination and that prevents water from reaching any of the sulfidic waste rock that will be produced by the project, either in a backfilled pit or in a waste rock dump.  The waste rock must be effectively sealed from water.  Otherwise, the project will cause unnecessary or undue degradation and the BLM will be violating its own 3809 regulations by allowing it.

 

Newmont as a parent company is responsible for the activities of Battle Mountain Gold

 

The BLM states that Newmont owns BMG as a wholly owned subsidiary and that they do not anticipate that the merger will affect the potential environmental impacts of the project.  FEIs at i.  This will be true only if Newmont takes responsibility for BMG’s actions.  If they hide behind the corporate veil such that only BMG is liable for the bond, then there is a significant potential problem.  In other words, Newmont must assume responsibility for the clean-up needed at the site.  Because Great Basin Mine Watch, and the EPA, believe that Newmont can not clean this site and operate the mine profitably and that the site will eventually be added to the National Priority List under the Superfund program, it is essential to document the degree of control that Newmont has over BMG. 

 

The Supreme Court recently considered the liability of parent corporations for the damages caused by their subsidiaries at sites that received Superfund cleanup.  See United States v. Best Foods, et al, 118 S.Ct. 1876 .  While setting a high bar for piercing the corporate veil, the Supreme Court found “A corporate parent that actively participated in, and exercised control over, the operations of its subsidiary's facility may be held directly liable in its own right under § 107(a)(2) [of CERCLA] as an operator of the facility”.  Best Foods at 1880.  It was not enough for the parent to hold majority stock or to control most or all of the subsidiary’s board of directors.  In this case, and for the record, as Newmont has told its shareholders and the Securities and Exchange Commission, Newmont actively controls BMG.

 

Newmont has made it obvious to its investors and the world that it intends to actively control BMG:

 

On January 10, 2001, Newmont completed a merger with Battle Mountain Gold Company pursuant to an agreement and plan of merger, dated June 21, 2000, under which each share of common stock of Battle Mountain and each exchangeable share of Battle Mountain Canada Ltd. (a wholly-owned subsidiary of Battle Mountain) was converted into the right to receive 0.105 share of NMC, or approximately 24.1 million shares.  Newmont also exchanged 2.3 million shares of newly issued $3.25 convertible preferred stock for all outstanding shares of Battle Mountain $3.25 conve3rtible preferred stock.  The merger will be accounted for as a pooling of interests, and as such, future consolidated financial statements will include Battle Mountain’s financial data as if Battle Mountain had always been part of Newmont.

 

Battle Mountain has operations in Canada and Bolivia, the Phoenix project in Nevada...The Phoenix project adds to Newmont’s domestic reserve base and will provide an opportunity for additional synergies in future years from utilization of existing nearby processing facilities.  Newmont’s 2000 Annual Report, at 17 and 18, emphases added.

 

Once again, for the record, Newmont is the operator of the Phoenix Project and will be liable for any costs that may be incurred in the future under Superfund liability.

 

Newmont recently acquired two new mining companies, Normandy and Franco-Nevada.  According to an article in the Financial Times, Newmont substantially overpaid for this acquisition.  It is our opinion that Newmont may have become financially unstable even though their stock prices have reached their highest levels in several years.  In response to our comment on Newmont’s financial stability, the BLM merely responded that Newmont was the world’s second largest gold producer.  In response to that, all we can say is “Remember Enron”.  Enron had many activities all over the world and the larger they became, the more unstable they became.  The BLM may not responsible for assessing the stability of project proponents, and on its face, Newmont appears better able than others to complete this project.  But to avoid the chance of leaving a legacy of extreme degradation at this site, the BLM must require better bonding and must establish that Newmont will be responsible for the activities[1].  The next section addresses reclamation requirements that have not been adequately discussed.  This is followed by a discussion on the financial surety aspects of the project.

 

 

Reclamation

 

This section addresses reclamation as proposed by Newmont and the BLM for this project.  The comments herein are supplemented by the technical review performed by Mr. Jim Kuipers.  His technical comments which expand on these statements are attached to this letter and should be considered to be wholly incorporated into this comment letter.  The reclamation plan and bonding calculations are not up to industry standards or even BLM standards as applied in other states for sites with significant AMD potential.  Mr. Kuipers notes that “[i]t is significant that up to the present time acid drainage has literally not been recognized or accounted for in reclamation and closure planning by the responsible state and federal regulatory agencies in Nevada.”.

 

The plan should be modified to recognize the following requirements:  At least 18 inches of suitable growth medium (salvaged topsoil to the greatest extent possible), including having the desired texture (density, coarse fragments and clay contents), organic matter, moisture retention, fertility, and pH characteristics adequate to propagate sustainable vegetation.  The plan is conspicuous in its lack of salvage of topsoil.  Other re-vegetation factors that also must be assessed include re-vegetation through time, slope and aspect, competition, weedy species and undesirable invader plants, and plant cover.  All these aspects should be included in the design and monitoring plan for the revegetation test plots and additional variables such as alternative cover designs and thickness, materials of construction, and revegetation approaches should also be evaluated.

 

The underlying assumption of the cost estimate for revegetation should consider the likelihood that successful revegetation will require a sustained rather than one time effort as depicted in the reclamation plan.  Experience has shown that multiple plantings of at least some areas is to be expected due to climatic, insect and other factors, and that fertilization over an extended time period is typically necessary, particularly where mined rock is to be substituted for a more suitable growth medium.

 

The need to eliminate infiltration into the acid-generating waste rock dumps, leach piles and tailing facilities, as well as the backfilled open pit, along with the type of vegetative treatment planned for reclamation, should dictate soil replacement depth.  In order to reduce infiltration to the greatest extent practical, a water balance cover (in which plants transpire water and the soil prevents water infiltration to underlying mine waste), or a water barrier cover (in which an impermeable water barrier to prevent water infiltration) should be specified for all potentially acid-generating areas.  The engineered cap should be designed so as to optimize desirable characteristics and achieve minimize infiltration to underlying acid generating waste while maximizing revegetation so as to increase evapotranspiration.

 

The FEIS has not adequately addressed tailings facility fluid management from the existing mine, proposed project, or post-reclamation standpoint, and it has not been recognized in corresponding financial assurance determinations.  The proposed plan should be modified to assume the most likely scenario of fluid management as being active management including capture and treatment due to the acid drainage generating nature of the tailings, the potential cost of which should also be included in the financial assurance determination.

 

The FEIS has not adequately addressed long-term heap leach fluid management and it has not been recognized in corresponding financial assurance determinations.  The proposed plan should be modified to assume the most likely scenario to fluid management as being active management including capture and treatment due to the acid drainage generating nature of the heap leach materials, the potential cost of which should also be included in the financial assurance determination.

 

The proposed plan does not address acid drainage that will likely result from the tailings facility and heap leach areas, as well as from the open pits and other areas.  It similarly does not address ground water contamination that is presently and in the future will continue to emanate from the existing tailings facilities.  These features make it obvious that there is an immediately existing as well as future need to conduct water management activities at the site that should be included in the reclamation and closure planning.  The requirements in this regard are likely to result in significant financial assurance requirements.

 

 

Financial Assurance

 

There are two aspects of financial assurance for this project.  One is the reclamation bond which allows for site closure should Newmont go bankrupt.  The second is the trust fund designed to monitor and mitigate long-term water quality problems likely to occur at the site.  Initially, we object very strongly to the BLM’s refusal to provide a copy of the Long-Term Contingency Fund.  That refusal was illegal in that it violated CEQ regulations that require that all referenced material in NEPA documents be available for public inspection.  “No material may be incorporated by reference unless it is reasonably available for inspection by potentially interested persons within the time allowed for comment. Material based on proprietary data which is itself not available for review and comment shall not be incorporated by reference.”  40CFR1502.21.  The BLM clearly referenced the document and then refused to make it available in violation of this regulation.  The refusal also violates the statement made by the BLM to the EPA in response to their concerns about this issue.

 

The U.S. Environmental Protection Agency (USEPA) and the public may review this financial information in the Phoenix Project Plan of Operations and all associated documents.  Specifically, the Reclamation Plan (Section 6 of the Pan of Operations) and the Phoenix Project Long-term Contingency Fund (Battle Mountain Gold 2001) discuss funding for reclamation and closure and postclosure monitoring and mitigation at the site.  Response to Comment 1-3, FEIS, emphases added.

 

This clearly states that the public would be allowed to review the Long-term Contingency Fund.  We clearly reserve the right to review this document and appeal its contents to the Interior Board of Land Appeals and federal court.  And we wonder whether the BLM intended to lie to the public or whether it was accidental.

 

Reclamation Bond:  The reclamation bond must include sufficient amounts to fully clean the site should BMG/Newmont go bankrupt at any time.  The plan includes backfill to prevent the formation of pit lakes; the financial assurance must include estimates for backfilling should bankruptcy occur at any time.

 

The FEIS (p. 2-48) provides an estimate for reclamation and fluid management of approximately $32,073,000 for the first 3-year phase of the proposed operation.  The maximum forecasted reclamation cost estimate for the project during any phase is approximately $55,800,000 during years 21-23 of operation.  As experience has shown at several mine sites including at the Zortman and Landusky, and Golden Sunlight mines, both of which are regulated by Montana BLM, the presence of acid generation can significantly impact financial assurance requirements.  AMD has resulted in an increase of approximately an order of magnitude to pay for surface reclamation and water treatment associated mitigation, and has raised the typical cost of reclamation and closure from an aggregated $5,000 - $15,000 per acre, to in excess of $50,000 to $100,000 or more per acre.

 

The Golden Sunlight Mine in Montana and the Molycorp Questa Mine in New Mexico are examples of where reclamation and closure plans requiring long-term water treatment have been decided and financial assurance established.  The two financial assurance requirements, at approximately $63 million and $152 million, respectively, are currently the largest financial assurances at operating mines in the U.S., and are typical of the result of the realization of acid drainage generating characteristics as they affect mine reclamation and closure planning requirements and financial assurance determination.  Additional information on both these mines has been attached.

 

The Zortman-Landusky Mine is another significant case in point with respect to bonding.  Despite the existence of significant bonding of approximately $72 million when its owner Pegasus Gold went bankrupt, the BLM and Montana DEQ have recently identified a need for an additional $33 million to address a shortfall for surface reclamation and water treatment in perpetuity for their preferred alternative.  The primary factors attributable to the financial assurance shortfall were an underestimate of the acid drainage generation at the site and a corresponding need for additional surface reclamation and water treatment capacity and long-term financial assurance.  In our opinion and as discussed throughout this comment set and the comment set on the DEIS, and in the FEIS itself, this site is as contaminated as these other sites.

 

Our review of the Phoenix Project financial assurance estimate shows that it has been appropriately performed as a first order engineering cost estimate.  Cost estimates for wages (Davis-Bacon), equipment costs (hourly rental costs from vendors), and adjusted unit costs are consistent with our experience and recommendations and the revised 3809 Surface Mining Regulations (effective January 2001).  We noted the following inconsistencies with recommended practice in the Phoenix Project cost estimate.

 

Direct Costs:  The direct costs in the reclamation cost estimate are for the proposed plan.  The previous sections comments on reclamation and closure identified deficiencies in the proposed plan with respect to cover design and materials, revegetation, and water treatment for existing as well as future ground water remediation.

 

Based on experience at other sites where reclamation and closure of similar mine sites having acid drainage has taken place, including those cited elsewhere in these comments, direct capital costs for surface reclamation and water management will most likely be higher than estimated for the proposed plan.  Engineered covers designed to minimize infiltration are likely to require either modified or different material characteristics, and it is likely that a more desirable growth medium for revegetation can be located in close proximity to the site.  The inadequacy of the proposed cap has been discussed in detail elsewhere.  Adequate revegetation establishment to promote evapotranspiration will likely require more intensive and longer-term effort than suggested by the proposed plan.  Capital requirements for water management including capture, conveyance, storage, treatment and discharge will most likely significantly exceed the estimated requirements in the proposed plan, both in terms of quantity of flow treated as well as load of acid and dissolved contaminants.

 

It is not possible to make an accurate estimate as to the actual required tasks and corresponding costs for reclamation and closure at the Phoenix Project site given the inadequacy of the information in the FEIS.  The FEIS is based upon a largely conceptual reclamation and closure plan that assumes a low to moderate level of activity and cost associated with acid generation.  However, cost estimates for comparable sites that are based on adequate information and identify and address acid generation are more complex in their treatment of the problem and have a significantly higher unit and total cost than in the cost estimate for the proposed plan.  Based on an average total unit cost of $75,000 per acre for reclamation at similar projects (see projects cited elsewhere in these comments and as provided in the Appendix) versus the $14,452 per acre in the cost estimate for the proposed plan, a shortfall of approximately $60,000 per acre or more than $100 million total may exist for the proposed bond amount.

 

The existence of this discrepancy in the financial assurance amount, given the similarity of the Phoenix Project’s characteristics to other mine sites where plans have been established that adequately deal with acid drainage and associated issues, supports the recommendation that a more extensive and immediate assessment of existing contamination and reclamation and closure requirements and associated financial liability be conducted by the responsible agencies, with assistance from the EPA and other knowledgeable agencies.  If any future mining takes place, it should be assessed from the standpoint of potentially aggravating a significant existing problem, and it similarly should be more adequately addressed before approval is given in order to ensure an adequate reclamation and closure plan and financial assurance.

 

Indirect Costs: The calculations inappropriately accounted for indirect costs including cost escalation and administrative costs.  The financial assurance estimate provided by the project proponent is based on a lump sum of costs in 2001 dollars.  Actual reclamation would take at least three to five years or more in which event the bond, which would be paid out as reclamation occurs unless otherwise specified, would not keep up with inflation and other factors.  It is standard practice to establish a reclamation schedule (see Golden Sunlight Bond Calculations in Appendix) and include cost escalation in order to determine the NPV of the financial assurance instrument for the year of its establishment.  A correction should be made to account for both escalation since the 2001 cost estimate was performed as well as future cost escalation.  The typical cost escalation factors used are 3.0% per year or as estimated by the Means Heavy Construction Costing Index (approximately 3.3% average per year over last ten years).

 

The Office of Surface Mining (OSM) has established long-standing costs associated with administrative overheads of agency operated reclamation and closure activities.  The OSM guidelines identify mobilization and demobilization, contingencies, engineering redesign, contractor profit and overhead, and reclamation management fees as indirect cost categories.  In addition individual states must include applicable insurance, tax and contractor bonding requirements.  The cost estimate for the proposed plan includes mobilization and demobilization at 5%, contingencies at 6%, no allowance for engineering redesign, contractor profit and overhead at 10%, and BLM administration at 10%.  In addition, Insurance (1.5% of labor) and performance and payment bonds (3%) were included.  The contractor overhead and profit should be corrected to 20% to reflect typical costs related to government contracting procedures, and an allowance of at least 5% should be made for engineering redesign.  This would result in an increase in indirect costs of approximately 15% over that contained in the cost estimate for the proposed plan.

 

The final cost estimate provided by Mr. Kuipers is that the bond must be at least $27,000,000 more than currently estimated.  This would be approximately $58,000,000 before the project may commence. 

 

The three-year financial assurance review needs to do more than adjust for additional area and cost escalation.  It should also be reviewed with respect to potential changes in site aspect or characterization and the reclamation and closure plan modified as necessary to mitigate future occurrences with corresponding changes made to the financial assurance estimate.  It should consider additional clean-up requirements as evidenced by changing groundwater quality conditions.  As pits are constructed, the costs of backfilling and liming that backfilling must be included[2].

 

 

Long-term Surety: Newmont will be required to establish a trust fund for treatment of groundwater pollution in the future.  The assumptions used in the proposed plan assumes short-term and long-term water management requirements for up to and beyond 130 years and greater than 1000 years, but in all cases requiring eventual groundwater capture and treatment at some point in the future for an extended period of time.  However, it would be erroneous to base the actual long-term water management costs on the estimates contained in the FEIS as it also states (p. 3.2-55) it is important to note that there is considerable uncertainty associated with long-term predictions of potential impacts to ground water quality resulting from infiltration through the waste rock facilities.  We discuss reasons for this uncertainty in detail at other parts of this letter.

 

Funding for perpetual treatment has been dealt with in a consistent matter at other sites in other states, including those where BLM was responsible, by requiring the financial assurance estimate to assume that all existing and eventual water treatment facilities were available either during or immediately post-mining and reclamation.  This is reflected in the cost estimates for the Golden Sunlight Mine and Molycorp Questa Mine reclamation and closure plans (attached to Jim Kuipers technical comments).

 

Using industry standard interest rates (3% inflation and 6% return on investment), Mr. Kuipers recalculated the present value of the contingency fund to be $9.1 million.  Because of the uncertainty in the calculations that will likely lead to significantly more contamination and treatment being needed much sooner, Mr. Kuipers calculated that $47.3 million should be placed in the trust fund today.  Kuipers’ calculations, which are attached to this letter, demonstrates the significant shortfall of the proposed long-term financial assurance based on the assumptions of the proposed plan, as well as the even greater shortfall which might occur due to reasonably expected acid drainage characteristics.  These figures highlight the gross inadequacy of the proposed financial assurance amount as well as the factors which have been used to calculate the amount.  It is recommended that the amount of financial assurance in the form of a trust fund for water treatment in perpetuity be calculated to include existing and future water treatment needs in the present and assumed to be necessary immediately following mining and reclamation, and that those calculations be based on conservative rather than best case scenarios.

 

Bonding for No-Action:  Currently, the site is not being mined.  The existing Reona project calls for mining for another six months, but that “would depend upon economic conditions”.  FEIS at ii.  Federal mining regulations require that reclamation begin at sites after mining has ceased for ____ months.  Because we doubt that Newmont/BMG can actually afford to mine this project under current economic conditions, we request that the BLM require reclamation to begin immediately if mining does not begin.  The FEIS also correctly states that for the No Action alternative, “the existing facilities identified in Section 2.2 would be closed and reclaimed in accordance with current permits and applicable federal and state closure and reclamation requirements”.  FEIS at 2-7.  Unfortunately, the current reclamation plan would allow the existing pits to “be left in their final mining configuration”.  It appears that the current reclamation plan would violate state law because it does not protect waters of the state.  The analysis of the pit lakes formed for the No Action alternative indicates that downgradient groundwater would be degraded by a pit lake in the existing Fortitude Pit.  The pit lake model indicates that the pit lake would have “some constituents exceeding secondary drinking water quality standards” and that there would be “an outflow of pit lake water to downgradient groundwater”.  FEIS at 3.2-74. 

 

Also, we request that the BLM immediately recalculate the bond required for the Reona Project, which, because of Newmont’s liability for full site clean-up, is essentially a bond needed for the No Action alternative as amended to include adequate clean-up.   Even for the No Action alternative, there may acidic water in the Minnie Pit.  “If water does pond in the Minnie Pit, it would likely be acidic with some elevated metals concentrations...”.  FEIS at 3.2-76.  The BLM’s assertion that the “spontaneous drainage of this pit indicates” the pit would likely “remain dry in the future” is unfounded.  Id.  As discussed elsewhere, the groundwater level in the area changes substantially and frequently and the pit would likely have water at least occasionally.  If the pit lake water is bad, its drainage will also degrade waters of the state.  Therefore, it is essential that even under No Action, backfill of this pit must include lime.  Therefore, it is essential to calculate a new bond for the existing condition to assure that Newmont actually fulfills its obligations to clean up the existing site. 

 

The BLM has argued that bonding for pit backfilling was not needed in the initial estimate.  On the DEIS, we commented that the reclamation bond was insufficient because it only applied to backfilling Minnie Pit.  Comment 13-38.   BLM’s response was that life-of-mine bonding was not necessary because the backfilling is only to “optimize materials handling”.  Because this pit will have significant water quality problems, backfilling and liming the Minnie Pit must be included in the initial bond calculations.

 

 

Water Resources

 

There are few places in the western United States where an active mine is so badly polluted as this site.  Even though the BLM documented this pollution in the DEIS, the BLM downplayed our comment regarding the current extent of pollution at the site.  Comment 13-4.  It must be the change in political leadership, but the BLM appears to now deny the pollution that exists at the site.  While the DEIS outlined many exceedences, in response to our noting of this fact, the BLM provided a previously ignored opinion that “pre-mining concentrations ... probably naturally exceeded current water quality standards”.   There may be some truth that the groundwater has naturally high constituent levels, but the current concentrations essentially qualify the site for Superfund status which is not natural.  Interestingly, the EPA agrees with us.  See comment 1-1.

 

We suggested that the BLM consider a project that had shallower pits because the technical waste rock report stated that acid generating potential increases with depth.  Comment 13-6.  In their response, the BLM ignores the technical report by stating that because the faults which control sulfide placement are vertical, a shallower project will not encounter less sulfide.  Why did the BLM ignore the waste rock report which indicated more sulfidization with depth?

 

Also, we requested that the information in Exponent Tables 7-1 and 7-2 be included in the EIS because it shows the long-term release of sulfate to the groundwater under the waste rock dumps, including backfilled open pits.  Comment 13-7.  The response is that it was left out of the DEIS because the BLM considered the predictions to be extremely conservative because the modeling was based on an assumption that all of the sulfide would oxidize and all of the resultant sulfate would be released to groundwater.  Unfortunately for the BLM’s assumption the model is conservative, Exponent stated that the model has three major components: “1)the oxidation of sulfide minerals in the waste rock produces a source of solutes, 2) chemical reactions between acidic pore waters and waste rock can attenuate the migration of some solutes, and 3)eventually, it is assumed that non-attenuated solutes are carried to the water table by meteoric water percolating through the waste rock.”  Exponent, page 47.  This is not conservative, it is just a model approach.  “After exhausting the neutralizing potential, sulfate produced by oxidation is assumed to move conservatively to the water table dissolved in the pore water, and loading will then decrease at a rate corresponding to the decrease in the sulfide oxidation rate; loading will cease when all the sulfide has been oxidized.”  Exponent, page 47 and 48.  These quotes from a longer discussion show that both the neutralizing capacity of the waste rock and the attenuation in underlying rock has been considered.

 

We also pointed out in our comments various physical realities that will likely increase the speed with which the sulfate will reach the groundwater.  Comment 13-8.  The BLM chose to downplay all of these realities by just saying that Exponent used reasonable estimates.  But Exponent did not test the sensitivity of their model to the uncertainty in their assumptions as any good modeler would do and the BLM did not request sensitivity analysis as any good reviewer would do.  In their classic text on groundwater modeling, Anderson and Woessner wrote “[t]he purpose of a sensitivity analysis is to quantify the uncertainty in the calibrated model caused by uncertainty in the estimates of aquifer parameters, stress, and boundary conditions.”[3]   They go on to state:

 

Not only do we have uncertainty as to the parameter values needed for our design calculations, we even have uncertainty about the very geometry of the system we are trying to analyze.  The uncertainties of lithology, stratigraphy, and structure introduce a level of complexity to geotechnical and hydrogeological analysis that is completely unknown in other engineering disciplines.[4]

 

While referring to saturated flow modeling, the principle of uncertainty clearly applies to flow modeling through the waste rock dumps.  Clearly, the BLM should require Exponent to test the sensitivity of their model with respect to parameters such as rooting depth and plant density.  Failure to do so means the BLM will fail to consider the uncertainty of the predictions.  For example, if the infiltration rate through the cap is just a little higher than predicted, it is likely that sulfate will reach the groundwater much quicker than predicted.  Without knowing the sensitivity of the model, the BLM has no basis for assuming the predictions are conservative.

 

Modeling of Flow through Waste Rock:  We also raised questions about the quality of one-dimensional modeling.  Comments 13-11 and 12.  The BLM’s response to 13-11 directed us to several excellent figures of moisture content with depth in existing waste rock dumps.  The BLM’s interpretation of the figures appears correct, but only if the assumption is made in advance to do one dimensional modeling.  In other words, the data has been interpreted to fit the chosen model rather than being analyzed with the idea of choosing the appropriate model.   For example, Figure A9-1c does show a general average moisture content of 4.5% at the surface dropping to about 1.5% at 50 feet as suggested by the BLM’s response.  However, one of the profiles shows a peak of 8% at about 40 feet (UFOISA-DH1) and another (NEOISA-DH2) shows a peak of 6% at about 35 feet.  This indicates slugs of water moving through the waste rock at different rates dependent on the location within the waste rock dump.  Other profiles show spikes which indicate slugs of water at different locations in the waste rock.  The line for CUOISA-DH1 shows a wetting front, but it is not distinct.  And, it trends toward very low values at depth.  This suggests that some water gets out in front of the wetting front, which is common in unsaturated flow.  In other words, if the wetting front model were accurate, the moisture content below the front would be relatively constant.  Instead, the figure suggests water moves downward at differing rates depending on the size and connectivity of pores.  This points to the need for three-dimensional modeling and the conclusion that the wetting fronts shown in Figures in other reports associated with this document are at best rough guess of where the water will be.

 

The response to 13-12 further indicates the need for three-dimensional modeling.  “Visual observations at this and other mines indicate that end-dumped waste rock facilities are typically heterogeneous and anisotropic, containing sloping layers of coarser and finer material.”  Response 13-12.  Treating this heterogeneous and anisotropic material as one-dimensional violates the assumptions of the models, namely that the material at least be heterogeneous.  We agree with the BLM that parameterizing a 3-d model of a waste rock dump would be difficult.  Data is available only where bore holes have been drilled.  For waste rock dumps yet to be constructed, it is only possible to consider existing data at other sites.  A desirable model for such a situation would likely employ stochastic simulation wherein the waste rock layers are divided into cells and the properties of each cell are entered as a frequency distribution.  The output from the model may still be a wetting front, but it will have a confidence band around it.  Instead of predicting it will take 130 years to reach the groundwater, it is probable that there is a chance that some water will reach the groundwater in just a few years.  With this kind of modeling, it would be possible to account for the uncertainty in the modeling.

 

The climatology input to the model is also incorrect.  Specifically, the 100-year rainfall is wrong (comment 13-13) and the scaling of the precipitation amount to higher elevations is incorrect (comment 13-14).  The BLM’s responses merely talk around the comments ignoring the comment’s substance.  While acknowledging the data showing that observed 24-hour precipitation values at nearby cities exceeds the 100-year estimate for Battle Mountain, they dismiss the concern by saying these values are consistent.  The difference of a few tenths of an inch can be very important because it may result in significant additional amounts of infiltration.  The first inch or so may just fill a deficit in the cap; additional amounts of precipitation likely add to the total that seeps through the cap to the waste rock. 

 

Response 13-14 makes no hydrologic sense.  The BLM’s basic argument is to agree with our comment (“More frequent precipitation in smaller events would tend to increase relative humidity and decrease evapotranspiration...”), but to deny it without any quantitative analysis ( “this effect would be offset, or even exceeded, by the fact that smaller precipitation events would produce less infiltration).  The response ignores reality.  One of the main factors limiting infiltration is that each precipitation event must fill the water deficit in the cap before seepage begins.  More frequent events keep the cap wetter; more of the larger events will actually seep into the waste rock.  And at higher elevations, this project spans 2000 feet, even the BLM realizes that lower evaporation occurs.  FEIS at 3.2-1.

 

Our comment 13-15 questioned the use of Battle Mountain solar radiation data at the higher elevations of this.  The BLM responded by saying that accounting for differences in elevation for precipitation is more important.  This is not correct.  With the magnitude of annual evaporation being near 40 inches while that of precipitation being around 10 inches, small variations, say 10%, in evaporation have a much large effect on the water balance of the cap than do similar magnitude variations in precipitation (1 inch).  Thus, saying that “infiltration is much less sensitive to these parameters than to precipitation” is just plain wrong.  To prove it, a sensitivity analysis of the model should be run.

 

Response 13-16 again shows a lack of understanding of the process.  We pointed out that the bedrock underlying the waste rock, or more specifically, the colluvium which underlies the waste rock, should be analyzed as a fractured media.  When considering it as a homogeneous bulk, the hydraulic conductivity, whether saturated or not, is very low, typically, and in this case, on the order of feet per year.  Assuming such homogeneity results in predictions that flow through the bedrock will take years to reach the groundwater.  In reality, when the bedrock underlies a media, such as colluvium or waste rock, that has a hydraulic conductivity orders of magnitude greater than the bedrock, the bedrock acts as a dam.  Water will literally pond on the surface reaching saturation relatively rapidly.  Once saturation is reached, the water will being to flow into the cracks or fractures.  (Capillary tension prevents the water from flowing from small pores into large fractures prior to reaching atmospheric pressure.)  Flow through fractures would literally be at rates of feet per hour.  In fractured rock, most of the flow is through the fractures.  Because the area of the fractures is very small as a function of the overall area of the bedrock, possibly covering less than 0.01% of the full cross-sectional area, the average (Darcy) velocity will be very small while the actual velocity (the rate the water flows through the actual pores) will be quite high.  Water, and contaminants, will reach the groundwater long before the average velocity would have predicted.  This is what is wrong with this seepage analysis.

 

In summary for the seepage analysis through waste rock, all of the comments raised by Great Basin Mine Watch suggest that flow to the groundwater will be much faster than determined by the BLM.  The fact that this pollution will occur much quicker than predicted changes the numbers used in the trust fund calculations as well as the reclamation plan.  See the discussion above regarding the funding of trust.  Hydrologic science indicates that the BLM is grossly ill-prepared to clean up the site should Newmont fail to do so and that Newmont is grossly underestimating the time until their liability begins.

 

Groundwater Levels: Figure 3.2-9 shows groundwater levels based on the wells show in Figure 3.2-8.  The level map shows contours in the upper elevations where there are no wells.  The document should have noted that there are no measurements of groundwater levels at high elevations.  Usually, springs are places where the groundwater level intersects the ground surface.

 

More importantly, the document fails to consider the screening levels of the wells in Figure 3.2-8.  The discussion fails to note whether artesian pressure has been observed.  It was observed during mine exploration in 1999 and 2000 and was used as an explanation for changing water levels in pits.

 

Pit Backfilling:  The FEIS fails to provide any additional information concerning the prevention of acid mine drainage from the waste rock to be backfilled into pit lakes.  BMG/Newmont will rely on submergence and added lime to prevent AMD.  However, the document does not specify how much lime will be needed.  Because this is likely to be an extremely large amount, it is essential that the FEIS discuss this because it may not even be available at a cost that will allow the mine to be profitably completed.  There may not be enough lime available from existing sources in the region.  The BLM needs to verify the amount and show that it can actually be obtained.  It must also be included in the reclamation bond calculations (see the section on reclamation bonding above.)

 

One of our primary concerns is that the groundwater levels rise and fall substantially in the backfilled pits and that this will allow oxygen to reach deep into the waste rock.  We raised this in comment 13-24.  The BLM relies on an analysis that the observed water level changes was caused by mine exploration.  First, our comment stated that the reports show a system “controlled by the annual amounts and seasonal distribution of recharge”, but the BLM apparently felt we said the entire 160 foot variation at some wells was dependent on seasonal recharge.  Comment 13-24.   Our statement clearly indicated annual variation as well.  The figures referenced by BLM (Figures 2-4A, 2-4B, and 2-4C in Baker) show significant declines over a several year period (trends) for wells CM35 (80 ft), CM38 (25 ft), CM41 (20 ft), CM53 (70 ft), CM30 (80 ft), CM40c (120 ft), CM41 (20 ft), CM44(130 ft), CM45 (100 ft), CP05 (160 ft) and CP06 (160 ft).  The majority, but not all, of the decline occurred during 1999.  In fact, the two wells cited by BLM with the largest decrease (CP05 and CP06) actually increased during preceding years by as much as 40 feet before commencing their decline.  The BLM ignored this increase.  Comparison of Figures 2-4A, B and C with Figure 1-2A in Baker (the location map) shows that the monitoring wells in the mountains fluctuated significantly and the wells on the flats south of the mine site remained relatively constant.  This would be expected as the lower elevation wells likely tap a regional aquifer into which smaller aquifers upgradient drain.  The further from the recharge zone, the more the water levels flatten out.  Unfortunately, Baker did not provide hydrographs of all the wells on Figure 1-2A which raises questions about their selection of wells.

 

There are many reasons to doubt Baker’s explanation for the groundwater decline between the North and South Midas pits.  A 160 foot decline over a 3/4 mile long and an estimated 1/4 wide region is 192 acre-feet (120 ac x160 ft x 0.01).  This is the approximate area of the coalescing drawdown around the two drawdown centers.  Baker at 2-18.  The storage coefficient is based on standard values for unconfined bedrock aquifers.  The calculation ignores the sloping sides of a drawdown cone.  Based on FEIS Table 3.2-17, this is at least 10% of the recharge in the entire basin.  The amounts measured by Baker emanating from exploration holes are only about 30% of this amount.  Baker page 2-16.  If the water levels fell because of exploration activities as described, there should have been concomitant rises in levels some where else.

 

Baker argues that the drill holes shown on Baker Figure 2-11 intercepted recharge flowing to monitoring wells CP05 and CP06 thereby explaining the level drop.  This would be an acceptable explanation if these wells produced an amount corresponding with the volume of the cone of depression and if the exploration wells were actually upgradient from the monitoring wells.  There also is no explanation of the increase in levels observed before the drop.  They are not upgradient as much as they are both on the eastern side slope of Copper Canyon.  While faults and fractures make the flow very complicated, it seems unlikely that these exploration wells could have intercepted flow heading toward these monitoring wells. 

 

Baker’s explanation that boreholes drilled “across the low permeability fault south of the South Midas Pit” may allow “leakage” that “provide[s] a continuing stress to the faulted and fractured bedrock system allowing ground water to leave the fractured aquifer and flow into the overlying alluvium” would be believable if there was evidence that flow was accumulating in the alluvium.  Baker at 2-17, 18.  There is no such evidence.

 

The BLM is incorrect to say that it is inappropriate to compare Sleeper with this site.  Comment 13-25.  It is an example where a similar plan, submerging sulfidic rock with water, failed to work as quickly as planned.  Another difference is that the company in that case had failed to predict the pit lake would be acidic.  It is conceivable that Newmont should do the same with one or more of the pit lakes that could form if backfilling is not completed.

 

The FEIS relies on pit backfilling without guaranteeing that backfill will not bury additional mineral resources.  Newmont has conducted condemnation drilling of beneficiation site facilities, but there is no discussion of requiring similar drilling before pits are backfilled.  FEIS at 2-47.

 

The BLM fails to explain how they will maintain security in the pits, in perpetuity.  Is there funding provided to maintain the proposed berms and warning signs?  FEIS at 2-46

 

Surface Discharge:  There is no explanation to our concern about the discharge of stormwater to surface drainages.  Comment 13-27.  In our comment, we mentioned that the EIS calls for discharging water to a point.  The BLM has not identified that point or otherwise addressed concerns about whether that point drains into Iron Canyon or other ephemeral or intermittent drainage.  The BLM completely ignored questions about the lining of the retention ponds and the fact that seepage from those ponds into a surface drainage would constitute an illegal discharge.

 

Surface Water: The primary stream of concern to be affected by this project is Willow Creek.  As described in the FEIS.  “[U]pper reaches of Willow Creek are in contact with the ground water system.  Gains in stream flow occur by net ground water inflow along the reach extending from the headwaters to a position on the local alluvial fan where it leaves the mountain front and begins to coalesce with a more extensive fan system.”  FEIS at 3.2-8.  This acknowledged the connection and therefore the dependence of this stream on the groundwater.  As we discuss on page 12-13 of our DEIS comments, it is illegal for the dewater to dry this stream.  The BLM claimed that we stretch the definition of the Clean Water Act and cited the original Clean Water Act.  Unfortunately, the BLM did not investigate the legal citations that used the CWA to prevent stream dewatering.

 

The BLM’s response to our comment regarding the degradation of springs and Public Water Reserve No.107 is incorrect.  Comment 13-33.  As explained in our comment, neither the reserve nor the IBLA decision allows for any mitigation by any method other than avoidance.   Thus, all of the springs listed on Table 3.2-14, regardless of the ownership of water rights, are potentially protected by this Public Water Reserve. 

 

 

Long-term Management Plan:  Our comments showed that the long-term management plan is insufficient to protect the groundwater.  The BLM failed to respond to the specifics of our comment 13-44 which pointed to the leading groundwater modeling text that stated that using MODFLOW to monitor fracture flows was a gross simplification of the model.  The fact that Baker was able to calibrate a porous media model to observed well levels for a fracture system is not surprising for a nonunique model[5].  Modeling contaminant transport without considering the fractures or other issues we raised is just wrong.  Contaminants will not flow through fractured systems like they flow through porous systems, the assumption made by using MODFLOW and discussed at length in our comment 13-44..  Unless there is a well monitoring every fracture system, the collection system will not collect, or even detect, the bulk of groundwater contamination.  The fact that MODFLOW has been used to “evaluate potential mine dewatering impacts to water levels in fractured rocks aquifers (sic) in numerous other mines in northern Nevada” is irrelevant to this discussion.  Those models would not be any more successful in predicting contaminant pathways than this one[6].   Also, those models all have uniqueness problems as well.  I refer you to any standard textbook on systems engineering for a discussion on uniqueness and how it affects modeling.  Singh[7] is a good example for hydrologic modeling.

 

As noted, we’ve expressed our concern that it is unlikely that any groundwater retrieval system based simply on wells will work.  We suggested in comment 13-45 the use of slurry walls and grout curtains.  We are and were aware of the difficulties cited in the BLM’s response that Newmont will encounter in installing these barriers.  Our comment indicated a need for “prescriptions [to] handle the uncertainties”.  Barriers should be used with the well retrieval system.  The barriers will cut-off fractures and divert the flow so that a larger proportion may be retrieved by the wells.  Ultimately, the amount of contaminations captured by the system will depend on the quality of the installation of the system.  It may well be that even with a combined barrier and well system, it may not be possible to collect enough of the contaminated water.  The BLM must complete a detailed assessment of this, including a reanalysis of the groundwater model, to determine whether Newmont can actually complete this project.  Perhaps the No Action alternative is the only one that will not degrade the groundwater.

 

Additional Review of the Long-term Contingency Plan: Because of the importance of this aspect of the project and because the BLM relies on the monitoring that occurs herein, we provide some additional review of the plan.

 

To be effective, wells must be placed in all fault and fracture zones.  While the plan’s objectives ignore fracture zones, the report does allude to the need to place wells in them.  CLGWM at 4 and 5.  Apparently, although it is not described, the authors used their automated MODFLOW calibration to identify places of high permeability.  The problem with this is that the scale of the fracture zones is much less than the scale of the model cells.

 

The authors rely on a five-foot layer of oxide waste rock as a cap over of the waste rock dumps to neutralize 30% of the acid to be generated.  CLGWM at 8 and 10.  Unfortunately, they do not identify the source of this neutralizing waste rock.  Neither does the FEIS.  Without specifying where the neutralizing waste rock will be found, it should not be planned on in the FEIS.

 

The modeling of contaminant particle transport has ignored dispersion.   For this reason, the particle tracking codes that utilize MODFLOW “are not appropriate when it is necessary to compute the first arrival of measurable contamination”.[8]

 

An additional poor choice was to assume that no oxidation occurs in the bedrock beneath the waste rock.  “Because most proposed waste-rock facilities are underlain by bedrock containing fractures with a low surface area to volume ratio, it was assumed that no additional attenuation of sulfate or acidity would be achieved once percolating water from the waste rock enters the bedrock system.”  CLGWM at 9.  Without proper documentation, which has not been provided, this assumption could have huge ramifications.   Fracture zones between the waste rock and the groundwater table could have high area to volume ratios.  Zones that have been dewatered could have experienced oxidation.  See Comment 15-6 by Glenn Miller.   There could be a significant increase in acidity as a result of this infiltration.  Also, there could be a significant increase in the leaching of metals from this rock.  Ignoring this without adequate data could be ignoring a large additional source of contamination.

 

Additionally, the modeling assumed that the lime added to open pits is completely successful in preventing further acidification.  Unfortunately, responses to our comments about the adequacy of the amendments included reference to the monitoring plan.  For example, see comment and response 13-24[9].  There is no explanation how a monitoring plan developed assuming there would be no release could actually monitor for such release.

 

The technical memorandum documenting the modeling indicates that existing waste rock on the site has already been neutralized.   “The predicted sulfate loading to groundwater is based on the assumption that the amount of acid produced during the brief period of oxidation in the existing waste rock is less than the neutralization potential of the material.  That is, once the old waste rock is covered with new waste rock and active oxidation stops, pore wastes return to near-neutral pH, and sulfate concentrations in the ore water in the old waste rock are limited by the solubility of gypsum”.  At 6.  Basically, the argument is that covering the existing waste rock will stop oxidation and the pH will become neutral.  There is no data or analysis provided to document this assumption.  Thus, it appears that the modeling may significantly underestimate the loading.

 

Figures 1 and 2, showing the location of extraction wells, somewhat confirms the issues we raised above concerning the lack of consideration of fracture zones.  If fractures were actually considered, the wells would not have been evenly spaced; few fracture zones are evenly spaced.  There is also no discussion of the depth or screening levels of the wells.  Most contaminants coming from the pits would be very deep and possibly in aquifers that are semi-confined and separate from upper layers.  Thus, pumping higher layers may not retrieve contaminants in lower levels.

 

The discussion in the FEIS constantly mentions this extraction plan and how it will protect the groundwater from all problems and the plan itself lays out neatly spaced wells based on isotropic and homogeneous conditions (in each model cell), allowing for nicely defined flow paths.  In conclusion, the analysis, modeling and design give this reviewer no confidence that the groundwater will actually be protected.  Newmont and the BLM, by placing all of their eggs into the monitoring and extraction basket, are establishing a legacy of pollution in the region that will last for centuries.

 

 

Groundwater Model:  The BLM at least noted most of our comments on the groundwater model without substantially disagreeing with them.  (Comments 13-62 through 13-80).  One exception is where we stated that mean error is a poor choice of calibration measures.  Comment 13-67.  Amazingly, the BLM responded by referring to page 238 of Anderson and Woessner, one of our common references.  Unfortunately, the book provides the formula and the following text (on page 240).  “The ME is simple to calculate but is usually not a wise choice because both negative and positive differences are incorporated in the mean and may cancel out the error.  Hence,